Openembedded Devel Discussions
 help / color / mirror / Atom feed
From: Stefan Schmidt <stefan@datenfreihafen.org>
To: openembedded-devel@lists.openembedded.org
Subject: Re: 2011.03 release testing, starts soon!
Date: Thu, 10 Feb 2011 09:31:33 +0100	[thread overview]
Message-ID: <20110210083133.GL28019@excalibur.local> (raw)
In-Reply-To: <AANLkTinrp6netE73j_iQmHu+sA5T4sUDEJJbijL45dKR@mail.gmail.com>

Hello.

On Wed, 2011-02-09 at 12:30, Khem Raj wrote:
> On Wed, Feb 9, 2011 at 2:18 AM, Koen Kooi <k.kooi@student.utwente.nl> wrote:
> > On 09-02-11 10:56, Stefan Schmidt wrote:
> >
> >>> 2. Do not delete the release branch after 2011.03 will be released (just like
> >>>    it was done for 2010.12), but let it live and allow developpers committing
> >>>    bug-fixes (backporting choosen things?) reported back by OE users (some would
> >>>    would be happy to contribute this way)
> >>
> >> That was already discussed. We make a tag with the release rev from which can be
> >> branched again _if_ people are stepping up to support this branch on a mid or
> >> long term base.
> >>
> >> The branch Tom is using until the release is pretty useless froma history point
> >> of view (all changes must be in master as well). When he thinks the release is
> >> good enough the tag gets added and the old branch deleted. For the last release
> >> nobody cared to support it afterwards with bugfixes so no release branch was
> >> created.
> >>
> >> I'm thinking about this for the upcoming release. If all works well we will base
> >> a product on it which I would like to support directly from such a release
> >> branch.
> >>
> >> The hard part is how people could decide on pooling resources on this. Defining
> >> goals for such a branch and stuff. E.g. only take serious fixes? What about
> >> package updates? Security fixes? changes on the toolchain or classes?
> >>
> 
> I would suggest to take only bugfixes and refrain from infrastructure
> changes which usually
> happen metadata wide and toolchain and classes are biggest part of
> this and with time backports
> into long term branches become more and more complex and painful and
> it may be that you have to do
> entirely different patch to solve the same issue in this branch which
> has been fixed differently upstream

Staying away from infrastructure and toolchain changes is something I really
want to see for such a branch.

From my BugLabs side I would expect bug fixes but also some package updates.
That would be stuff based on the java/osgi framework we are deploying so
hopefully it would not make problems for anyone else. I don't expect to much
problems with that.

With regards to the lifetime I don't see this as a several years branch fo our
product (thinking about oe-core for the next release). Obviously it is still
fine for other people to use it that long. The current stable branch is a good
example for this. It is still used and different companies/people are happy with
it having no updates at all for some time. :)

I would expect something else for the branch on top of the 2011-03 release. Some
more changes in the begining while fixing problems that poped up after the tag
was created but commits will slow down after some months.

> I think if such a branch is to be created it should be created from
> the release tag after the release happens

Agreed.

> >> This is up to the group who wants to support such a branch. Anyone else
> >> interested in doing this for 2011-03?
> 
> If such a branch is expected for long term then its better to decide
> its policies on patches
> from upstream pov all fixes that go into this branch should first go
> into master if its not
> a direct backport of some sort. That way we can be sure that we dont
> lose changes that should be in master.

Agreed as well. It gets a mess if we start fixing stuff in the release branch
only.

regards
Stefan Schmidt



  reply	other threads:[~2011-02-10  8:32 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-02-08 19:41 2011.03 release testing, starts soon! Tom Rini
2011-02-09  9:25 ` Aeschbacher, Fabrice
2011-02-09  9:56   ` Stefan Schmidt
2011-02-09 10:18     ` Koen Kooi
2011-02-09 10:42       ` Stefan Schmidt
2011-02-09 11:20         ` Koen Kooi
2011-02-09 20:30       ` Khem Raj
2011-02-10  8:31         ` Stefan Schmidt [this message]
2011-02-09 14:47   ` Tom Rini
2011-02-10  9:42     ` Aeschbacher, Fabrice

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20110210083133.GL28019@excalibur.local \
    --to=stefan@datenfreihafen.org \
    --cc=openembedded-devel@lists.openembedded.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox